All of lore.kernel.org
 help / color / mirror / Atom feed
* [Intel-gfx] [PATCH] drm/i915: Fix DP clock recovery "voltage_tries" handling
@ 2021-10-01 16:08 Ville Syrjala
  2021-10-01 17:17 ` Imre Deak
                   ` (2 more replies)
  0 siblings, 3 replies; 4+ messages in thread
From: Ville Syrjala @ 2021-10-01 16:08 UTC (permalink / raw)
  To: intel-gfx; +Cc: Imre Deak, Jani Nikula

From: Ville Syrjälä <ville.syrjala@linux.intel.com>

The DP spec says:
"If the receiver keeps the same value in the ADJUST_REQUEST_LANEx_y
 register(s) while the LANEx_CR_DONE bits remain unset, the transmitter
 must loop four times with the same voltage swing. On the fifth time,
 the transmitter must down-shift to the lower bit rate and must repeat
 the CR-lock training sequence as described below."

Lets fix the code to follow that instead of terminating after five
times of transmitting the same signal levels. The text in spec feels
a little bit ambiguous still, but this is my best guess at its meaning.

As a bonus this also gets rid of the train_set[0] stuff which
would not work for per-lane drive settings anyway.

Cc: Imre Deak <imre.deak@intel.com>
CC: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
---
 .../drm/i915/display/intel_dp_link_training.c | 29 +++++++++++++++----
 1 file changed, 24 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 053ed9302cda..73a823f1ec22 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -511,6 +511,25 @@ static void intel_dp_link_training_clock_recovery_delay(struct intel_dp *intel_d
 		drm_dp_lttpr_link_train_clock_recovery_delay();
 }
 
+static bool intel_dp_adjust_request_changed(int lane_count,
+					    const u8 old_link_status[DP_LINK_STATUS_SIZE],
+					    const u8 new_link_status[DP_LINK_STATUS_SIZE])
+{
+	int lane;
+
+	for (lane = 0; lane < lane_count; lane++) {
+		u8 old = drm_dp_get_adjust_request_voltage(old_link_status, lane) |
+			drm_dp_get_adjust_request_pre_emphasis(old_link_status, lane);
+		u8 new = drm_dp_get_adjust_request_voltage(new_link_status, lane) |
+			drm_dp_get_adjust_request_pre_emphasis(new_link_status, lane);
+
+		if (old != new)
+			return true;
+	}
+
+	return false;
+}
+
 /*
  * Perform the link training clock recovery phase on the given DP PHY using
  * training pattern 1.
@@ -521,7 +540,7 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp,
 				      enum drm_dp_phy dp_phy)
 {
 	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
-	u8 voltage;
+	u8 old_link_status[DP_LINK_STATUS_SIZE] = {};
 	int voltage_tries, cr_tries, max_cr_tries;
 	bool max_vswing_reached = false;
 
@@ -574,8 +593,6 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp,
 			return false;
 		}
 
-		voltage = intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK;
-
 		/* Update training set as requested by target */
 		intel_dp_get_adjust_train(intel_dp, crtc_state, dp_phy,
 					  link_status);
@@ -585,12 +602,14 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp,
 			return false;
 		}
 
-		if ((intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK) ==
-		    voltage)
+		if (!intel_dp_adjust_request_changed(crtc_state->lane_count,
+						     old_link_status, link_status))
 			++voltage_tries;
 		else
 			voltage_tries = 1;
 
+		memcpy(old_link_status, link_status, sizeof(link_status));
+
 		if (intel_dp_link_max_vswing_reached(intel_dp, crtc_state))
 			max_vswing_reached = true;
 
-- 
2.32.0


^ permalink raw reply related	[flat|nested] 4+ messages in thread
end of thread, other threads:[~2021-10-02  5:27 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-01 16:08 [Intel-gfx] [PATCH] drm/i915: Fix DP clock recovery "voltage_tries" handling Ville Syrjala
2021-10-01 17:17 ` Imre Deak
2021-10-01 22:13 ` [Intel-gfx] ✓ Fi.CI.BAT: success for " Patchwork
2021-10-02  5:27 ` [Intel-gfx] ✓ Fi.CI.IGT: " Patchwork

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.